Ethernet and Fast Ethernet Adapters
Adapter Specifications
Network adapters (NIC, Network Interface Card) Ethernet and Fast Ethernet can interface with a computer through one of standard interfaces:
- ISA (Industry Standard Architecture) bus;
- PCI bus (Peripheral Component Interconnect);
- PC Card bus (aka PCMCIA);
Adapters designed for the ISA system bus (backbone) were not so long ago the main type of adapters. The number of companies producing such adapters was large, which is why the devices of this type were the cheapest. Adapters for ISA are available in 8- and 16-bit. 8-bit adapters are cheaper, while 16-bit adapters are faster. True, information exchange on the ISA bus cannot be too fast (in the limit - 16 MB/s, in reality - no more than 8 MB/s, and for 8-bit adapters - up to 2 MB/s). Therefore, Fast Ethernet adapters, which require high transfer rates for effective operation, are practically not produced for this system bus. The ISA bus is becoming a thing of the past.
The PCI bus has now practically replaced the ISA bus and is becoming the main expansion bus for computers. It provides 32- and 64-bit data exchange and has high throughput (theoretically up to 264 MB/s), which fully satisfies the requirements of not only Fast Ethernet, but also the faster Gigabit Ethernet. It is also important that the PCI bus is used not only in IBM PC computers, but also in PowerMac computers. In addition, it supports Plug-and-Play automatic hardware configuration. Apparently, in the near future PCI bus the majority will be oriented network adapters. The disadvantage of PCI compared to the ISA bus is that the number of expansion slots in a computer is usually small (usually 3 slots). But exactly network adapters connect to PCI first.
The PC Card bus (old name PCMCIA) is currently used only in Notebook class portable computers. In these computers, the internal PCI bus is usually not routed to the outside. The PC Card interface allows for easy connection of miniature expansion cards to a computer, and the exchange speed with these cards is quite high. However, more and more laptop computers are equipped with built-in network adapters, as network connectivity becomes an integral part of the standard feature set. These onboard adapters are again connected to the computer's internal PCI bus.
When choosing network adapter oriented to a particular bus, you must first of all make sure that there are free expansion slots for this bus in the computer connected to the network. You should also evaluate the complexity of installing the purchased adapter and the prospects for producing boards of this type. The latter may be needed if the adapter fails.
Finally, they meet again network adapters, connecting to a computer via a parallel (printer) LPT port. The main advantage of this approach is that you do not need to open the computer case to connect adapters. In addition, in this case, adapters do not occupy computer system resources, such as interrupt channels and DMAs, as well as memory addresses and I/O devices. However, the speed of information exchange between them and the computer in this case is much lower than when using the system bus. In addition, they require more processor time to communicate with the network, thereby slowing down the computer.
Recently, there are more and more computers in which network adapters built into the system board. The advantages of this approach are obvious: the user does not have to buy a network adapter and install it in the computer. You just need to connect network cable to the external connector of the computer. However, the disadvantage is that the user cannot select the adapter with the best characteristics.
Other important characteristics network adapters can be attributed:
- adapter configuration method;
- size installed on the board buffer memory and modes of exchange with it;
- the ability to install a permanent memory chip for remote booting (BootROM) on the board.
- the ability to connect the adapter to different types of transmission media (twisted pair, thin and thick coaxial cable, fiber optic cable);
- the network transmission speed used by the adapter and the availability of its switching function;
- the adapter can use full-duplex exchange mode;
- compatibility of the adapter (more precisely, the adapter driver) with the network software used.
User configuration of the adapter was used primarily for adapters designed for the ISA bus. Configuration involves setting up the use of computer system resources (input/output addresses, interrupt channels and direct memory access, buffer memory addresses and remote boot memory). Configuration can be carried out by setting switches (jumpers) to the desired position or using the DOS configuration program supplied with the adapter (Jumperless, Software configuration). When starting such a program, the user is prompted to set the hardware configuration using a simple menu: select adapter parameters. The same program allows you to make self-test adapter The selected parameters are stored in the adapter's non-volatile memory. In any case, when choosing parameters, you must avoid conflicts with system devices computer and with other expansion cards.
The adapter can also be configured automatically in Plug-and-Play mode when the computer is turned on. Modern adapters usually support this particular mode, so they can be easily installed by the user.
In the simplest adapters, exchange with the internal buffer memory of the adapter (Adapter RAM) is carried out through the address space of input/output devices. In this case, no additional configuration of memory addresses is required. The base address of buffer memory operating in shared memory mode must be specified. It is assigned to the computer's upper memory area (
Today it is almost impossible to find a laptop or motherboard without an integrated network card, or even two. All of them have the same connector - RJ45 (more precisely, 8P8C), but the speed of the controller can differ by an order of magnitude. In cheap models it is 100 megabits per second (Fast Ethernet), in more expensive ones it is 1000 (Gigabit Ethernet).
If your computer does not have a built-in LAN controller, then it is most likely already an “old man” based on a processor like Intel Pentium 4 or AMD Athlon XP, as well as their “ancestors”. Such “dinosaurs” can be “made friends” with a wired network only by installing a discrete network card with a PCI connector, since the PCI Express bus did not yet exist at the time of their birth. But also for the PCI bus (33 MHz), “network cards” are produced that support the most current Gigabit Ethernet standard, although its throughput may not be enough to fully unleash the speed potential of a gigabit controller.
But even if you have a 100-megabit integrated network card, those who are going to “upgrade” to 1000 megabits will have to purchase a discrete adapter. The best option would be to buy a PCI Express controller, which will ensure maximum network speed, if, of course, the corresponding connector is present in the computer. True, many will prefer a PCI card, since they are much cheaper (the cost starts from literally 200 rubles).
What advantages will the transition from Fast Ethernet to Gigabit Ethernet bring in practice? How different real speed data transfer of PCI versions of network cards and PCI Express? Is normal speed enough? hard drive to fully load a gigabit channel? You will find answers to these questions in this material.
Test participants
The three cheapest discrete network cards (PCI - Fast Ethernet, PCI - Gigabit Ethernet, PCI Express - Gigabit Ethernet) were selected for testing, since they are in greatest demand.
The 100-megabit network PCI card is represented by the Acorp L-100S model (price starts from 110 rubles), which uses the Realtek RTL8139D chipset, the most popular for cheap cards.
The 1000-megabit network PCI card is represented by the Acorp L-1000S model (price starts from 210 rubles), which is based on the Realtek RTL8169SC chip. This is the only card with a heatsink on the chipset - the rest of the test participants do not require additional cooling.
The 1000-megabit network PCI Express card is represented by the TP-LINK TG-3468 model (price starts from 340 rubles). And it was no exception - it is based on the RTL8168B chipset, which is also produced by Realtek.
Appearance of the network card
Chipsets from these families (RTL8139, RTL816X) can be seen not only on discrete network cards, but also integrated on many motherboards.
The characteristics of all three controllers are shown in the following table:
Show table
The bandwidth of the PCI bus (1066 Mbit/s) should theoretically be enough to “boost” gigabit network cards to full speed, but in practice it may still not be enough. The fact is that this “channel” is shared by all PCI devices; in addition, it transmits service information on servicing the bus itself. Let's see if this assumption is confirmed by real speed measurements.
Another nuance: the vast majority of modern hard drives have an average reading speed of no more than 100 megabytes per second, and often even less. Accordingly, they will not be able to fully load the gigabit channel of the network card, the speed of which is 125 megabytes per second (1000: 8 = 125). There are two ways to get around this limitation. The first is to combine a pair of such hard drives into a RAID array (RAID 0, striping), and the speed can almost double. The second is to use SSD drives, the speed parameters of which are significantly higher than those of hard drives.
Testing
A computer with the following configuration was used as a server:
- processor: AMD Phenom II X4 955 3200 MHz (quad-core);
- motherboard: ASRock A770DE AM2+ ( AMD chipset 770 + AMD SB700);
- RAM: Hynix DDR2 4 x 2048 GB PC2 8500 1066 MHz (dual-channel mode);
- video card: AMD Radeon HD 4890 1024 MB DDR5 PCI Express 2.0;
- network card: Realtek RTL8111DL 1000 Mbit/s (integrated on the motherboard);
- operating system: Microsoft Windows 7 Home Premium SP1 (64-bit version).
A computer with the following configuration was used as a client into which the tested network cards were installed:
- processor: AMD Athlon 7850 2800 MHz (dual core);
- motherboard: MSI K9A2GM V2 (MS-7302, AMD RS780 + AMD SB700 chipset);
- RAM: Hynix DDR2 2 x 2048 GB PC2 8500 1066 MHz (dual-channel mode);
- video card: AMD Radeon HD 3100 256 MB (integrated into the chipset);
- HDD: Seagate 7200.10 160 GB SATA2;
- operating system: Microsoft Windows XP Home SP3 (32-bit version).
Testing was carried out in two modes: reading and writing via network connection from hard drives (this should show that they may be " bottleneck"), as well as from RAM disks in the RAM of computers that imitate fast SSD drives. The network cards were connected directly using a three-meter patch cord (eight-core twisted pair cable, category 5e).
Data transfer rate (hard drive - hard drive, Mbit/s)
The actual data transfer speed through the 100-megabit Acorp L-100S network card fell just short of the theoretical maximum. But both gigabit cards, although they outperformed the first by about six times, were unable to show the maximum possible speed. It is clearly seen that the speed is limited by the performance of Seagate 7200.10 hard drives, which, when directly tested on a computer, averages 79 megabytes per second (632 Mbit/s).
In this case, there is no fundamental difference in speed between network cards for the PCI bus (Acorp L-1000S) and PCI Express (TP-LINK), the slight advantage of the latter can be explained by the measurement error. Both controllers were operating at about sixty percent of their capacity.
Data transfer rate (RAM disk - RAM disk, Mbit/s)
Acorp L-100S expectedly showed the same low speed when copying data from high-speed RAM disks. This is understandable - the Fast Ethernet standard has not corresponded to modern realities for a long time. Compared to the “hard drive-to-hard drive” testing mode, the Acorp L-1000S gigabit PCI card significantly increased performance - the advantage was approximately 36 percent. The TP-LINK TG-3468 network card showed an even more impressive lead - the increase was about 55 percent.
This is where the higher bandwidth of the PCI Express bus showed itself - it outperformed the Acorp L-1000S by 14 percent, which can no longer be attributed to an error. The winner fell slightly short of the theoretical maximum, but the speed of 916 megabits per second (114.5 Mb/s) still looks impressive - this means that you will have to wait almost an order of magnitude less for the copying to complete (compared to Fast Ethernet). For example, the time to copy a 25 GB file (a typical HD rip with good quality) from computer to computer will be less than four minutes, and with a previous generation adapter it will take more than half an hour.
Testing has shown that Gigabit Ethernet network cards have a huge advantage (up to tenfold) over Fast Ethernet controllers. If your computers only have hard drives that are not combined into a striping array (RAID 0), then there will be no fundamental difference in speed between PCI and PCI Express cards. Otherwise, as well as when using high-performance SSD drives, preference should be given to cards with PCI interface Express, which will provide the highest possible data transfer speed.
Naturally, it should be taken into account that other devices in the network “path” (switch, router...) must support the Gigabit Ethernet standard, and the category of twisted pair (patch cord) must be at least 5e. Otherwise, the real speed will remain at 100 megabits per second. By the way, backward compatibility with the Fast Ethernet standard, it remains the same: you can connect, for example, a laptop with a 100-megabit network card to a gigabit network; this will not affect the speed of other computers on the network.
The ComputerPress testing laboratory tested Fast Ethernet network cards for the PCI bus intended for use in 10/100 Mbit/s workstations. The currently most common cards with a throughput of 10/100 Mbit/s were selected, since, firstly, they can be used in Ethernet, Fast Ethernet and mixed networks, and, secondly, the promising Gigabit Ethernet technology ( bandwidth up to 1000 Mbit/s) is still most often used to connect powerful servers to the network equipment of the network core. It is extremely important what quality of passive network equipment (cables, sockets, etc.) is used in the network. It is well known that if for Ethernet networks a twisted pair cable of category 3 is sufficient, then category 5 is already required for Fast Ethernet. Signal scattering and poor noise immunity can significantly reduce throughput networks.
The purpose of testing was to determine, first of all, the effective performance index (Performance/Efficiency Index Ratio - hereinafter P/E index), and only then - the absolute value of throughput. The P/E index is calculated as the ratio of the network card throughput in Mbit/s to the CPU load as a percentage. This index is the industry standard for measuring network adapter performance. It was introduced in order to take into account the use of CPU resources by network cards. The fact is that some network adapter manufacturers try to achieve maximum performance by using more computer processor cycles to perform network operations. Minimum processor load and relatively high throughput are essential for running mission-critical business, multimedia, and real-time applications.
We tested the cards that are currently most often used for workstations in corporate and local networks:
- D-Link DFE-538TX
- SMC EtherPower II 10/100 9432TX/MP
- 3Com Fast EtherLink XL 3C905B-TX-NM
- Compex RL 100ATX
- Intel EtherExpress PRO/100+ Management
- CNet PRO-120
- NetGear FA 310TX
- Allied Telesyn AT 2500TX
- Surecom EP-320X-R
The main characteristics of the tested network adapters are given in Table. 1 . Let us explain some of the terms used in the table. Automatic connection speed detection means that the adapter itself determines the maximum possible operating speed. In addition, if auto-speed detection is supported, no additional configuration is required when moving from Ethernet to Fast Ethernet and back. That is, from system administrator There is no need to reconfigure the adapter or reload drivers.
Support for Bus Master mode allows you to transfer data directly between the network card and computer memory. This frees up the central processor to perform other operations. This property has become a de facto standard. It’s no wonder that all well-known network cards support Bus Master mode.
Remote turn-on (Wake on LAN) allows you to turn on your PC over a network. That is, it becomes possible to service the PC during non-working hours. For this purpose, three-pin connectors are used on system board and a network adapter, which are connected with a special cable (included in delivery). In addition, special control software is required. Wake on LAN technology was developed by the Intel-IBM alliance.
Full duplex mode allows you to transmit data simultaneously in both directions, half duplex - only in one direction. Thus, the maximum possible throughput in full duplex mode is 200 Mbit/s.
The DMI (Desktop Management Interface) makes it possible to obtain information about the configuration and resources of a PC using network management software.
Support for the WfM (Wired for Management) specification ensures the interaction of the network adapter with network management and administration software.
To remotely boot a computer OS over a network, network adapters are equipped with special memory BootROM. This allows diskless workstations to be used effectively on a network. Most of the tested cards only had a BootROM slot; The BootROM chip itself is usually a separately ordered option.
ACPI (Advanced Configuration Power Interface) support helps reduce power consumption. ACPI is a new technology that powers the power management system. It is based on the use of both hardware and software. Basically, Wake on LAN is integral part ACPI.
Proprietary performance tools allow you to increase the efficiency of your network card. The most famous of them are Parallel Tasking II from 3Com and Adaptive Technology from Intel. These products are usually patented.
Support for major operating systems is provided by almost all adapters. The main operating systems include: Windows, Windows NT, NetWare, Linux, SCO UNIX, LAN Manager and others.
The level of service support is assessed by the availability of documentation, a floppy disk with drivers and the ability to download latest versions drivers from the company website. Packaging also plays an important role. From this point of view, the best, in our opinion, are the network adapters from D-Link, Allied Telesyn and Surecom. But overall the level of support turned out to be satisfactory for all cards.
Typically, the warranty covers the entire life of the AC adapter (lifetime warranty). Sometimes it is limited to 1-3 years.
Testing methodology
All tests used the latest versions of network card drivers, which were downloaded from the Internet servers of the respective manufacturers. In the case where the network card driver allowed any settings and optimization, the default settings were used (except for the Intel network adapter). Note that the richest additional features and functions are provided by cards and corresponding drivers from 3Com and Intel.
Performance measurements were performed using Novell's Perform3 utility. The principle of operation of the utility is that a small file is copied from the workstation to a shared one network drive server, after which it remains in the server’s file cache and is read from there many times over a given period of time. This allows for memory-network-memory interoperability and eliminates the impact of latency associated with disk operations. The utility parameters include initial file size, final file size, resizing step, and testing time. Novell Perform3 utility displays performance values with files different sizes, average and maximum performance (in KB/s). The following parameters were used to configure the utility:
- Initial file size - 4095 bytes
- Final file size - 65,535 bytes
- File increment step - 8192 bytes
The testing time with each file was set to twenty seconds.
Each experiment used a pair of identical network cards, one running on the server and the other running on the workstation. This seems to be inconsistent with common practice, as servers typically use specialized network adapters that come with a number of additional features. But this is exactly the way - the same network cards are installed on both the server and workstations - testing is carried out by all well-known test laboratories in the world (KeyLabs, Tolly Group, etc.). The results are somewhat lower, but the experiment turns out to be clean, since only the analyzed network cards work on all computers.
Compaq DeskPro EN client configuration:
- Pentium II 450 MHz processor
- cache 512 KB
- RAM 128 MB
- hard drive 10 GB
- operating system Microsoft Windows NT Server 4.0 c 6 a SP
- TCP/IP protocol.
Compaq DeskPro EP server configuration:
- Celeron processor 400 MHz
- RAM 64 MB
- hard drive 4.3 GB
- operating system Microsoft Windows NT Workstation 4.0 c c 6 a SP
- TCP/IP protocol.
Testing was carried out in conditions where the computers were connected directly with a UTP Category 5 crossover cable. During these tests, the cards operated in 100Base-TX Full Duplex mode. In this mode, the throughput is slightly higher due to the fact that part of the service information (for example, confirmation of reception) is transmitted simultaneously with useful information, the volume of which is estimated. Under these conditions, it was possible to record fairly high throughput values; for example, for the 3Com Fast EtherLink XL 3C905B-TX-NM adapter, the average is 79.23 Mbps.
CPU load was measured on the server using Windows utilities NT Performance Monitor; the data was recorded in a log file. The Perform3 utility was run on the client so as not to affect the server's processor load. The server computer processor was an Intel Celeron, whose performance is significantly lower than the performance of Pentium II and III processors. Intel Celeron was used deliberately: the fact is that since the processor load is determined with a fairly large absolute error, in the case of large absolute values the relative error is smaller.
After each test, the Perform3 utility places the results of its work in a text file in the form of a data set of the following form:
65535 bytes. 10491.49 KBps. 10491.49 Aggregate KBps. 57343 bytes. 10844.03 KBps. 10844.03 Aggregate KBps. 49151 bytes. 10737.95 KBps. 10737.95 Aggregate KBps. 40959 bytes. 10603.04 KBps. 10603.04 Aggregate KBps. 32767 bytes. 10497.73 KBps. 10497.73 Aggregate KBps. 24575 bytes. 10220.29 KBps. 10220.29 Aggregate KBps. 16383 bytes. 9573.00 KBps. 9573.00 Aggregate KBps. 8191 bytes. 8195.50 KBps. 8195.50 Aggregate KBps. 10844.03 Maximum KBps. 10145.38 Average KBp.
It displays the file size, the corresponding throughput for the selected client and for all clients (in this case there is only one client), as well as the maximum and average throughput for the entire test. The obtained average values for each test were converted from KB/s to Mbit/s using the formula:
(KB x 8)/1024,
and the P/E index value was calculated as the ratio of throughput to processor load as a percentage. Subsequently, the average value of the P/E index was calculated based on the results of three measurements.
The following problem arose when using the Perform3 utility on Windows NT Workstation: in addition to writing to a network drive, the file was also written to the local file cache, from where it was subsequently read very quickly. The results were impressive, but unrealistic, since there was no data transfer as such over the network. In order for applications to treat shared network drives as normal local disks, the operating system uses a special network component - a redirector, which redirects I/O requests over the network. Under normal operating conditions, when performing the procedure of writing a file to a shared network drive, the redirector uses the Windows NT caching algorithm. That is why when writing to the server, writing also occurs to the local file cache of the client machine. And to carry out testing, it is necessary that caching be carried out only on the server. To ensure that there is no caching on the client computer, Windows registry NT, the parameter values were changed, which made it possible to disable caching performed by the redirector. Here's how it was done:
- Path to Registry:
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Rdr\Parameters
Parameter name:
UseWriteBehind enables write-behind optimization for files being written
Type: REG_DWORD
Value: 0 (default: 1)
- Path to Registry:
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Lanmanworkstation\parameters
Parameter name:
UtilizeNTCaching specifies whether the redirector will use the Windows NT cache manager to cache file contents.
Type: REG_DWORD Value: 0 (Default: 1)
Intel EtherExpress PRO/100+Management Network Adapter
This card's throughput and CPU utilization were found to be almost the same as the 3Com's. The settings windows for this card are shown below.
New Intel controller The 82559 installed on this card provides very high performance, especially on Fast Ethernet networks.
The technology that Intel uses in its Intel card EtherExpress PRO/100+, named Adaptive Technology. The essence of the method is to automatically change the time intervals between Ethernet packets depending on the network load. As network congestion increases, the distance between individual Ethernet packets dynamically increases, which reduces the number of collisions and increases throughput. When the network load is light, when the probability of collisions is low, the time intervals between packets are reduced, which also leads to increased performance. The greatest benefits of this method should be seen in large collision Ethernet segments, that is, in cases where the network topology is dominated by hubs rather than switches.
New Intel technology, called Priority Packet, allows traffic passing through the network card to be regulated according to the priorities of individual packets. This makes it possible to increase data transfer rates for mission-critical applications.
Provides support for VLANs (IEEE 802.1Q standard).
There are only two indicators on the board - work/connection, speed 100.
www.intel.com
Network adapter SMC EtherPower II 10/100 SMC9432TX/MP
The architecture of this card uses two promising technologies: SMC SimulTasking and Programmable InterPacket Gap. The first technology is similar to 3Com Parallel Tasking technology. By comparing the test results for cards from these two manufacturers, we can draw a conclusion about the degree of effectiveness of the implementation of these technologies. We also note that this network card showed the third result both in terms of performance and P/E index, ahead of all cards except 3Com and Intel.
There are four LED indicators on the card: speed 100, transmission, connection, duplex.
The company's main website address is: www.smc.com
Introduction
The purpose of creating this report was a brief and accessible presentation of the basic principles of operation and features of computer networks, using Fast Ethernet as an example.
A network is a group of connected computers and other devices. The main purpose of computer networks is sharing resources and the implementation of interactive communication both within one company and beyond its borders. Resources are data, applications and peripherals, such as an external drive, printer, mouse, modem or joystick. The concept of interactive communication between computers implies the exchange of messages in real time.
There are many sets of standards for data transmission in computer networks. One of the sets is the Fast Ethernet standard.
From this material you will learn about:
- · Fast Ethernet technologies
- Switches
- FTP cable
- Connection types
- · Topologies computer network
In my work, I will show the principles of operation of a network based on the Fast Ethernet standard.
Local switching computer networks(LAN) and Fast Ethernet technologies were developed in response to the need to improve the efficiency of Ethernet networks. By increasing throughput, these technologies can eliminate network bottlenecks and support data-intensive applications. The appeal of these solutions is that you don't have to choose one or the other. They are complementary, so network efficiency can often be improved by using both technologies.
The collected information will be useful both to people starting to study computer networks and to network administrators.
1. Network diagram
2. Fast Ethernet technology
computer network fast ethernet
Fast Ethernet is the result of the development of Ethernet technology. Based on and retaining the same CSMA/CD (channel polling multiple access and collision detection) technique, Fast Ethernet devices operate at 10 times the speed of Ethernet. 100 Mbps. Fast Ethernet provides sufficient bandwidth for applications such as computer-aided design and manufacturing (CAD/CAM), graphics and image processing, and multimedia. Fast Ethernet is compatible with 10 Mbps Ethernet, so it's easier to integrate Fast Ethernet into your LAN using a switch rather than a router.
Switch
Using switches many workgroups can be connected to form a large LAN (see Diagram 1). Inexpensive switches perform better than routers, providing better LAN performance. Fast Ethernet workgroups consisting of one or two hubs can be connected through a Fast Ethernet switch to further increase the number of users as well as cover a larger area.
As an example, consider the following switch:
Rice. 1 D-Link-1228/ME
The DES-1228/ME series of switches includes premium, configurable Layer 2 Fast Ethernet switches. With advanced functionality, DES-1228/ME devices are inexpensive solution to create a secure and high-performance network. Distinctive Features The features of this switch are high port density, 4 Gigabit Uplink ports, small step change settings for bandwidth management, and improved network management. These switches allow you to optimize your network both in terms of functionality and cost characteristics. The DES-1228/ME series switches are optimal solution both in terms of functionality and cost characteristics.
FTP cable
Cable LAN-5EFTP-BL consists of 4 pairs of single-core copper conductors.
Conductor diameter 24AWG.
Each conductor is encased in HDPE (High Density Polyethylene) insulation.
Two conductors twisted with a specially selected pitch make up one twisted pair.
The 4 twisted pairs are wrapped in polyethylene film and, together with a single-core copper ground conductor, are enclosed in a common foil shield and PVC sheath.
Straight through
It serves:
- 1. To connect a computer to a switch (hub, switch) via the computer’s network card
- 2. To connect network peripheral equipment - printers, scanners - to the switch (hub, switch)
- 3. for UPLINK to a higher switch (hub, switch) - modern switches can automatically configure the inputs in the connector for reception and transmission
Crossover
It serves:
- 1. For direct connection of 2 computers to a local network, without the use of switching equipment (hubs, switches, routers, etc.).
- 2. for uplink, connection to a higher-level switch in a local network with a complex structure, for older types of switches (hubs, switches), they have a separate connector, also marked “UPLINK” or an X.
Star topology
To the stars- the basic topology of a computer network in which all computers on the network are connected to a central node (usually a switch), forming a physical segment of the network. Such a network segment can function either separately or as part of a complex network topology (usually a “tree”). All information exchange takes place exclusively through central computer, on which a very large load is placed in this way, so it cannot do anything else except the network. As a rule, it is the central computer that is the most powerful, and it is on it that all functions for managing the exchange are assigned. In principle, no conflicts in a network with a star topology are possible, because management is completely centralized.
Application
Classic 10 Mbit Ethernet suited most users for about 15 years. However, in the early 90s, its insufficient capacity began to be felt. For computers on Intel processors 80286 or 80386 s ISA buses(8 MB/s) or EISA (32 MB/s) Ethernet segment bandwidth was 1/8 or 1/32 of the memory-to-disk channel, and this was well consistent with the ratio of data volumes processed locally and data transferred over networks. For more powerful client stations with a PCI bus (133 MB/s), this share dropped to 1/133, which was clearly not enough. As a result, many 10Mbps Ethernet segments became overloaded, server responsiveness dropped significantly, and collision rates increased significantly, further reducing usable throughput.
There is a need to develop a “new” Ethernet, that is, a technology that would be equally cost-effective with a performance of 100 Mbit/s. As a result of searches and research, experts were divided into two camps, which ultimately led to the emergence of two new technologies - Fast Ethernet and l00VG-AnyLAN. They differ in the degree of continuity with classic Ethernet.
In 1992, a group of network equipment manufacturers, including Ethernet technology leaders such as SynOptics, 3Com and several others, formed the Fast Ethernet Alliance, a non-profit association, to develop a standard new technology, which was supposed to preserve the features of Ethernet technology to the maximum extent possible.
The second camp was led by Hewlett-Packard and AT&T, which offered to take advantage of the opportunity to address some of the known shortcomings of Ethernet technology. After some time, these companies were joined by IBM, which contributed by proposing to provide some compatibility with the new technology. Token networks Ring.
At the same time, IEEE Committee 802 formed a research group to study the technical potential of new high-speed technologies. Between late 1992 and late 1993, the IEEE team studied 100-Mbit solutions offered by various vendors. Along with the Fast Ethernet Alliance proposals, the group also reviewed high-speed technology proposed by Hewlett-Packard and AT&T.
The discussion centered on the issue of maintaining the random CSMA/CD access method. The Fast Ethernet Alliance proposal preserved this method and thereby ensured continuity and consistency between 10 Mbps and 100 Mbps networks. The HP-AT&T coalition, which had the support of significantly fewer vendors in the networking industry than the Fast Ethernet Alliance, proposed completely new method access named Demand Priority- priority access on demand. It significantly changed the behavior of nodes on the network, so it could not fit into Ethernet technology and the 802.3 standard, and a new IEEE 802.12 committee was organized to standardize it.
In the fall of 1995, both technologies became IEEE standards. The IEEE 802.3 committee adopted the Fast Ethernet specification as the 802.3 standard, which is not a standalone standard, but is an addition to the existing 802.3 standard in the form of chapters 21 to 30. The 802.12 committee adopted the l00VG-AnyLAN technology, which uses a new Demand Priority access method and supports two frame formats - Ethernet and Token Ring.
v Physical layer of Fast Ethernet technology
All differences between Fast Ethernet technology and Ethernet are concentrated on the physical layer (Fig. 3.20). The MAC and LLC layers in Fast Ethernet remain exactly the same and are described in the previous chapters of the 802.3 and 802.2 standards. Therefore, when considering Fast Ethernet technology, we will study only a few options for its physical layer.
The more complex structure of the physical layer of Fast Ethernet technology is due to the fact that it uses three types of cabling systems:
- · fiber optic multimode cable, two fibers are used;
- · Category 5 twisted pair, two pairs are used;
- · Category 3 twisted pair, four pairs are used.
Coaxial cable, which gave the world the first Ethernet network, was not included in the list of permitted data transmission media of the new Fast Ethernet technology. This is a common trend in many new technologies, because over short distances, Category 5 twisted pair cable can transmit data at the same speed as coaxial cable, but the network is cheaper and easier to operate. Over long distances, optical fiber has much higher bandwidth than coax, and the cost of the network is not much higher, especially when you consider the high troubleshooting costs of a large coaxial cable system.
Differences between Fast Ethernet technology and Ethernet technology
Refusal coaxial cable led to the fact that Fast Ethernet networks always have a hierarchical tree structure built on hubs, just like l0Base-T/l0Base-F networks. The main difference between Fast Ethernet network configurations is the reduction in network diameter to approximately 200 m, which is explained by a 10-fold reduction in the minimum length frame transmission time due to a 10-fold increase in transmission speed compared to 10 Mbit Ethernet.
Nevertheless, this circumstance does not really hinder the construction of large networks using Fast Ethernet technology. The fact is that the mid-90s were marked not only by the widespread use of inexpensive high-speed technologies, but also by the rapid development of local networks based on switches. When using switches, the Fast Ethernet protocol can operate in full-duplex mode, in which there are no restrictions on the total length of the network, but only restrictions on the length of the physical segments connecting neighboring devices (adapter - switch or switch - switch). Therefore, when creating long-distance local network backbones, Fast Ethernet technology is also actively used, but only in the full-duplex version, in conjunction with switches.
IN this section A half-duplex version of the Fast Ethernet technology is being considered, which fully complies with the definition of the access method described in the 802.3 standard.
Compared to the physical implementation options for Ethernet (and there are six of them), in Fast Ethernet the differences between each option and the others are deeper - both the number of conductors and coding methods change. And since the physical variants of Fast Ethernet were created simultaneously, and not evolutionarily, as for Ethernet networks, it was possible to define in detail those sublayers of the physical layer that do not change from variant to variant, and those sublayers that are specific to each variant of the physical environment.
The official 802.3 standard established three different specifications for the Fast Ethernet physical layer and gave them the following names:
Fast Ethernet Physical Layer Structure
- · 100Base-TX for two-pair cable on unshielded twisted pair UTP category 5 or shielded twisted pair STP Type 1;
- · 100Base-T4 for four-pair UTP Category 3, 4 or 5 UTP cable;
- · 100Base-FX for multimode fiber optic cable, two fibers are used.
The following statements and characteristics are true for all three standards.
- · Fast Ethernetee technology frame formats are different from 10 Mbit Ethernet technology frame formats.
- · The interframe interval (IPG) is 0.96 µs and the bit interval is 10 ns. All timing parameters of the access algorithm (backoff interval, minimum frame length transmission time, etc.), measured in bit intervals, remained the same, so no changes were made to the sections of the standard relating to the MAC level.
- · A sign of a free state of the medium is the transmission of the Idle symbol of the corresponding redundant code (and not the absence of signals, as in the 10 Mbit/s Ethernet standards). The physical layer includes three elements:
- o reconciliation sublayer;
- o media independent interface (Media Independent Interface, Mil);
- o physical layer device (PHY).
The negotiation layer is needed so that the MAC layer, designed for the AUI interface, can work with the physical layer through the MP interface.
The physical layer device (PHY) consists, in turn, of several sublayers (see Fig. 3.20):
- · logical data encoding sublevel, which converts bytes coming from the MAC level into 4B/5B or 8B/6T code symbols (both codes are used in Fast Ethernet technology);
- · physical connection sublayers and physical media dependence (PMD) sublayers, which provide signal generation in accordance with a physical coding method, for example NRZI or MLT-3;
- · autonegotiation sublayer, which allows two communicating ports to automatically select the most efficient operating mode, for example, half-duplex or full-duplex (this sublayer is optional).
The MP interface supports a medium-independent way of exchanging data between the MAC sublayer and the PHY sublayer. This interface is similar in purpose to the AUI interface of classic Ethernet, except that the AUI interface was located between the physical signal coding sublayer (for all cable options the same physical coding method was used - Manchester code) and the physical connection sublayer to the medium, and the MP interface is located between the MAC sublayer and signal coding sublevels, of which there are three in the Fast Ethernet standard - FX, TX and T4.
The MP connector, unlike the AUI connector, has 40 pins, the maximum length of the MP cable is one meter. Signals transmitted via the MP interface have an amplitude of 5 V.
Physical layer 100Base-FX - multimode fiber, two fibers
This specification defines the operation of the Fast Ethernet protocol over multimode fiber in half-duplex and full-duplex modes based on the well-proven FDDI encoding scheme. As in the FDDI standard, each node is connected to the network by two optical fibers coming from the receiver (R x) and from the transmitter (T x).
There are many similarities between the l00Base-FX and l00Base-TX specifications, so properties common to the two specifications will be given under the generic name l00Base-FX/TX.
While 10 Mbps Ethernet uses Manchester encoding to represent data over a cable, the Fast Ethernet standard defines a different encoding method - 4V/5V. This method has already proven its effectiveness in the FDDI standard and has been transferred without changes to the l00Base-FX/TX specification. In this method, every 4 bits of MAC sublayer data (called symbols) are represented by 5 bits. The redundant bit allows potential codes to be applied by representing each of the five bits as electrical or optical pulses. The existence of prohibited symbol combinations allows erroneous symbols to be rejected, which increases the stability of networks with l00Base-FX/TX.
To separate the Ethernet frame from the Idle characters, a combination of the Start Delimiter characters (a pair of characters J (11000) and K (10001) of the 4B/5B code is used, and after the completion of the frame, a T character is inserted before the first Idle character.
Continuous data flow of 100Base-FX/TX specifications
Once the 4-bit chunks of MAC codes are converted into 5-bit chunks of the physical layer, they need to be represented as optical or electrical signals in the cable connecting the network nodes. The l00Base-FX and l00Base-TX specifications use different physical encoding methods for this - NRZI and MLT-3, respectively (as in FDDI technology when working via optical fiber and twisted pair).
Physical layer 100Base-TX - twisted pair DTP Cat 5 or STP Type 1, two pairs
The l00Base-TX specification uses UTP Category 5 cable or STP Type 1 cable as the data transmission medium. The maximum cable length in both cases is 100 m.
The main differences from the l00Base-FX specification are the use of the MLT-3 method for transmitting signals of 5-bit portions of 4V/5V code over twisted pair, as well as the presence of an Auto-negotiation function for selecting the port operating mode. The autonegotiation scheme allows two physically connected devices that support several physical layer standards, differing in bit speed and number of twisted pairs, to select the most advantageous operating mode. Typically, the auto-negotiation procedure occurs when you connect a network adapter, which can operate at speeds of 10 and 100 Mbit/s, to a hub or switch.
The Auto-negotiation scheme described below is the l00Base-T technology standard today. Before this, manufacturers used various proprietary schemes automatic detection speeds of communicating ports that were not compatible. The Auto-negotiation scheme adopted as a standard was originally proposed by National Semiconductor under the name NWay.
A total of 5 different operating modes are currently defined that can support l00Base-TX or 100Base-T4 devices on twisted pairs;
- · l0Base-T - 2 pairs of category 3;
- l0Base-T full-duplex - 2 pairs of category 3;
- · l00Base-TX - 2 pairs of category 5 (or Type 1ASTP);
- · 100Base-T4 - 4 pairs of category 3;
- · 100Base-TX full-duplex - 2 pairs of category 5 (or Type 1A STP).
The l0Base-T mode has the lowest priority in the negotiation process, and the full-duplex 100Base-T4 mode has the highest. The negotiation process occurs when the device is turned on, and can also be initiated at any time by the device control module.
The device that has started the auto-negotiation process sends a packet of special impulses to its partner Fast Link Pulse burst (FLP), which contains an 8-bit word encoding the proposed interaction mode, starting with the highest priority supported by the node.
If the peer node supports the auto-negotuiation function and can also support the proposed mode, it responds with a burst of FLP pulses in which it confirms the given mode, and this ends the negotiation. If the partner node can support a lower priority mode, then it indicates it in the response, and this mode is selected as the working mode. Thus, the highest priority common node mode is always selected.
A node that only supports l0Base-T technology sends Manchester pulses every 16 ms to check the integrity of the line connecting it to a neighboring node. Such a node does not understand the FLP request that a node with the Auto-negotiation function makes to it, and continues to send its pulses. A node that receives only line integrity pulses in response to an FLP request understands that its partner can only operate using the l0Base-T standard, and sets this operating mode for itself.
Physical layer 100Base-T4 - twisted pair UTP Cat 3, four pairs
The 100Base-T4 specification was designed to allow high-speed Ethernet to use existing Category 3 twisted pair wiring. This specification increases overall throughput by simultaneously carrying bit streams over all 4 pairs of cable.
The 100Base-T4 specification appeared later than other Fast Ethernet physical layer specifications. The developers of this technology primarily wanted to create physical specifications closest to those of l0Base-T and l0Base-F, which operated on two data lines: two pairs or two fibers. To implement work over two twisted pairs, I had to switch to a higher quality Category 5 cable.
At the same time, the developers of the competing technology l00VG-AnyLAN initially relied on working over twisted pair cable of category 3; the most important advantage was not so much the cost, but the fact that it was already installed in the vast majority of buildings. Therefore, after the release of the l00Base-TX and l00Base-FX specifications, the developers of Fast Ethernet technology implemented their own version of the physical layer for twisted pair category 3.
Instead of 4V/5V encoding, this method uses 8V/6T encoding, which has a narrower signal spectrum and, at a speed of 33 Mbit/s, fits into the 16 MHz band of category 3 twisted pair cable (when encoding 4V/5V, the signal spectrum does not fit into this band) . Every 8 bits of MAC level information are encoded by 6 ternary symbols, that is, numbers that have three states. Each ternary digit has a duration of 40 ns. The group of 6 ternary digits is then transmitted onto one of the three transmit twisted pairs, independently and sequentially.
The fourth pair is always used to listen to the carrier frequency for collision detection purposes. The data transfer rate on each of the three transmit pairs is 33.3 Mbps, so the total speed of the 100Base-T4 protocol is 100 Mbps. At the same time, due to the adopted coding method, the signal change rate on each pair is only 25 Mbaud, which allows the use of category 3 twisted pair.
In Fig. Figure 3.23 shows the connection between the MDI port of a 100Base-T4 network adapter and the MDI-X port of a hub (the prefix X indicates that for this connector, the receiver and transmitter connections are exchanged in cable pairs compared to the network adapter connector, which makes it easier to connect pairs of wires in the cable - without crossing). Pair 1 -2 always required to transfer data from MDI port to MDI-X port, pair 3 -6 - to receive data by the MDI port from the MDI-X port, and the pair 4 -5 And 7 -8 are bidirectional and are used for both reception and transmission, depending on the need.
Connection of nodes according to the 100Base-T4 specification
Let us note the main features of the development of Ethernet networks and the transition to Fast Ethernet networks (IEEE 802.3u standard):
- - tenfold increase in throughput;
- - saving the CSMA/CD random access method;
- - saving frame format;
- - support for traditional data transmission media.
These properties, as well as support for two speeds and auto-sensing of 10/100 Mbit/s, built into Fast Ethernet network cards and switches, allow for a smooth transition from Ethernet networks to higher-speed Fast Ethernet networks, providing advantageous continuity compared to other technologies. Another additional factor for successfully conquering the market is the low cost of Fast Ethernet equipment.
Fast Ethernet architecture
The structure of the Fast Ethernet layers (including the MII interface and the Fast Ethernet transceiver) is shown in Fig. 13. Even at the development stage of the 100Base-T standard, the IEEE 802.3u committee determined that there is no universal signal encoding scheme that would be ideal for all three physical interfaces (TX, FX, T4). When compared with the Ethernet standard, the encoding function (Manchester code) is performed by the physical signaling layer PLS (Fig. 5), which is located above the medium-independent interface AUI. In the Fast Ethernet standard, encoding functions are performed by the PCS encoding sublayer, located below the media-independent MII interface. As a result, each transceiver must use its own set of encoding schemes best suited for the corresponding physical interface, such as the 4V/5V and NRZI set for the 100Base-FX interface.
MII interface and Fast Ethernet transceivers. The MII (medium independent interface) interface in the Fast Ethernet standard is analogous to the AUI interface in the Ethernet standard. The MII interface provides communication between the negotiation and physical encoding sublayers. Its main purpose is to simplify use different types environment. The MII interface requires further connection of a Fast Ethernet transceiver. A 40-pin connector is used for communication. The maximum distance along the MII interface cable should not exceed 0.5 m.
If the device has standard physical interfaces (e.g. RJ-45), then the physical layer sublayer structure can be hidden within the chip with large logic integration. In addition, deviations in the protocols of intermediate sublevels in a single device are acceptable, with the main goal of increasing performance.
Fast Ethernet physical interfaces
The Fast Ethernet IEEE 802.3u standard establishes three types of physical interface (Fig. 14, Table 6 Main characteristics of the physical interfaces of the Fast Ethernet IEEE 802.3u standard): 100Base-FX, 100Base-TX and 100Base-T4.
100Base-FX. The standard of this fiber optic interface is completely identical to the FDDI PMD standard. The main optical connector of the 100Base-FX standard is Duplex SC. The interface allows a duplex communication channel.
- * - the distance is achieved only in duplex communication mode.
- 100Base-TX. The standard of this physical interface requires the use of unshielded twisted pair cable of category no lower than 5. It is completely identical to the FDDI UTP PMD standard. The physical RJ-45 port, as in the 10Base-T standard, can be of two types: MDI (network cards, workstations) and MDI-X (Fast Ethernet repeaters, switches). A single MDI port may be present on a Fast Ethernet repeater.
For transmission over copper cable, pairs 1 and 3 are used. Pairs 2 and 4 are free. The RJ-45 port on the network card and on the switch can support, along with 100Base-TX mode, 10Base-T mode, or the auto-sensing function. Most modern network cards and switches support this function via RJ-45 ports and, in addition, can operate in full duplex mode.
100Base-T4. This type of interface allows you to provide a half-duplex communication channel over twisted pair UTP cat. 3 and above. It is the ability of an enterprise to migrate from the Ethernet standard to the Fast Ethernet standard without radically replacing the existing cable system based on UTP cat.3 that should be considered the main advantage of this standard.
Unlike the 100Base-TX standard, which uses only two twisted pairs of cable for transmission, the 100Base-T4 standard uses all four pairs. Moreover, when connecting the workstation and the repeater via a direct cable, data from the workstation to the repeater goes through twisted pairs 1, 3 and 4, and in the opposite direction - through pairs 2, 3 and 4. Pairs 1 and 2 are used for collision detection similar to the Ethernet standard . The other two pairs 3 and 4 alternately, depending on the commands, can pass the signal in either one or the other direction. Transmitting a signal in parallel over three twisted pairs is equivalent to inverse multiplexing, discussed in Chapter 5. The bit rate per channel is 33.33 Mbit/s.
Character coding 8V/6T. If Manchester encoding were used, the bit rate per twisted pair would be 33.33 Mbps, which would exceed the 30 MHz limit for such cables. An effective reduction in the modulation frequency is achieved if, instead of a direct (two-level) binary code, a three-level (ternary) code is used. This code is known as 8B/6T; this means that before transmission occurs, each set of 8 binary bits (character) is first converted according to certain rules into 6 triple (three-level) symbols.
The 100Base-T4 interface has one significant drawback - the fundamental impossibility of supporting duplex transmission mode. And if, when building small Fast Ethernet networks using repeaters, 100Base-TX has no advantages over 100Base-T4 (there is a collision domain, the bandwidth of which is no more than 100 Mbit/s), then when building networks using switches, the disadvantage of the 100Base-T4 interface becomes obvious and very serious. Therefore, this interface has not become as widespread as 100Base-TX and 100Base-FX.
Fast Ethernet Device Types
The main categories of devices used in Fast Ethernet are the same as in Ethernet: transceivers; converters; network cards (for installation on workstations/file servers); repeaters; switches.
Transceiver- a two-port device covering the PCS, PMA, PMD and AUTONEG sublevels, and having, on the one hand, an MII interface, and on the other, one of the environment-dependent physical interfaces (100Base-FX, 100Base-TX or 100Base-T4). Transceivers are used relatively rarely, just as network cards, repeaters, and switches with an MII interface are rarely used.
LAN card. The most widely used network cards today are those with a 100Base-TX interface to the PCI bus. Optional, but highly desirable, features of the RJ-45 port are 100/10 Mbps auto-configuration and full duplex support. Most modern cards being released support these features. Network cards with a 100Base-FX optical interface are also available (manufacturers IMC, Adaptec, Transition Networks, etc.) - the main standard optical one is the SC connector (ST is allowed) for multimode OB.
Converter(media converter) is a two-port device, both ports of which represent media-dependent interfaces. Converters, unlike repeaters, can operate in full duplex mode, except when there is a 100Base-T4 port. 100Base-TX/100Base-FX converters are common. Due to the general trends in the growth of broadband long-distance networks using single-mode fiber optics, the consumption of optical transceivers for single-mode fiber optics has increased sharply in the last decade. Converter chassis that combine multiple individual 100Base-TX/100Base-FX modules allow you to connect multiple fiber segments converging in a central hub to a switch equipped with full-duplex RJ-45 (100Base-TX) ports.
Repeater. Based on the parameter of maximum time delays when relaying frames, Fast Ethernet repeaters are divided into two classes:
- - Class I. RTD dual run latency should not exceed 130 watts. Due to less stringent requirements, repeaters of this class can have T4 and TX/FX ports, and can also be stacked.
- - Class II. Repeaters of this class are subject to more stringent requirements for double-path delay: RTD
Switch - important device corporate networks. Most modern Fast Ethernet switches support 100/10 Mbps auto-configuration on RJ-45 ports and can provide full-duplex communication on all ports (except 100Base-T4). Switches may have special additional slots for installing an up-link module. The interfaces for such modules can be optical ports such as Fast Ethernet 100Base-FX, FDDI, ATM (155 Mbit/s), Gigabit Ethernet, etc.
Large switch manufacturers Fast Ethernet companies are: 3Com, Bay Networks, Cabletron, DEC, Intel, NBase, Cisco, etc.
Loading...
